Semiquantitative analysis using whole-body dynamic F-18 fluoro-2-deoxy-glucose-positron emission tomography to differentiate between benign and malignant lesions.

Abstract

To investigate whether whole-body dynamic positron emission tomography (PET) is useful for differentiating benign and malignant lesions. In this retrospective study, data from a cohort of 146 lesions from 187 patients who consecutively underwent whole-body dynamic PET scans at our hospital for suspected lesions in the lung, lymph nodes, liver, bone, esophagus, and colon were analyzed. Patients with malignant lymphomas, accumulations > 5cm in length along the long axis of the esophagus, or lesions in the colon in which the site of accumulation moved during the imaging period were excluded. Patients were administered 3.7MBq/kg of fluorine-18-fluorodeoxyglucose (F-18 FDG), and dynamic imaging was initiated 60min after administration. We defined the 60-65, 65-70, 70-75, and 75-80min time mark as the first, second, third, and fourth pass, respectively. The static image is the summed average of all the four pass images. We measured the accumulation in the mean image of the whole-body dynamic PET scan, which was arithmetically similar to the maximum standardized uptake value (SUVmax) throughout the whole-body static images obtained during 20min of imaging (S-SUVmax). The ratio of SUVmax in the dynamic first pass(60-65min after FDG administration) and fourth pass(75-80min after FDG administration) was calculated as R-SUVmax. The S-SUVmax in the lung, lymph nodes, and bone did not differ significantly between the benign and malignant groups. However, there was a significant difference in R-SUVmax, which was > 1 in most malignant lesions indicating an increase in accumulation during routine scan time. Significant differences were observed between benign and malignant lesions of the liver in both S-SUVmax and R-SUVmax values, with the latter being > 1 in most malignant lesions. Whole-body dynamic PET for 20min starting 1h after FDG administration improved the accuracy of malignant lesion detection in the liver, lymph nodes, lung, and bone. The incremental improvement was small, and the FDG dynamics in the distribution of values between benign and malignant overlapped. Additional information from whole-body dynamic imaging can help detect malignant lesions in these sites without increasing patient burden or prolonging imaging time.